Can Codestral 21B Pruned i1 run on Gaudi 3 128GB?

YES — Runs Great

C46Usable

○Estimated from fit model

Codestral 21B Pruned i1 needs ~29.0 GB VRAM. Gaudi 3 128GB has 128.0 GB. With Q4_K_M quantization, expect ~202 tok/s.

Runtime: llama.cppCapacity: RoomyBandwidth: HighStack: StandardBottleneck: Balanced

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 29.0 GB, 202.2 tok/s, Runs well

29.0 GB required128.0 GB available

23% VRAM used

Fit status

Runs well

Decode

202.2 tok/s

TTFT

958 ms

Safe context

660K

Memory

29.0 GB / 128.0 GB

Memory breakdown

Weights12.8 GB

KV Cache2.5 GB

Runtime0.9 GB

Headroom12.8 GB

See how fast it feels

See how fast it feelsCodestral 21B Pruned i1 on Gaudi 3 128GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:

def quicksort(arr):
 if len(arr) <= 1:
 return arr
 pivot = arr[len(arr) // 2]
 left = [x for x in arr if x < pivot]
 middle = [x for x in arr if x == pivot]
 right = [x for x in arr if x > pivot]
 return quicksort(left) + middle + quicksort(right)

# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers)) # [1, 1, 2, 3, 6, 8, 10]

The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
 """Sort a list using the quicksort algorithm.

 Args:
 arr: List of integers to sort.

 Returns:
 A new sorted list.
 """
 if len(arr) <= 1:
 return arr
 pivot = arr[len(arr) // 2]
 left = [x for x in arr if x < pivot]
 middle = [x for x in arr if x == pivot]
 right = [x for x in arr if x > pivot]
 return quicksort(left) + middle + quicksort(right)

Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 202.2 tok/s decode · 958ms TTFT (warm) · 506 tok/s prefill

What limits this setup

The raw memory story may look fine, but the software ecosystem is still a constraint here.

Runtime ecosystem is narrower than CUDA

Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.

Best improvement path

Prefer CUDA if you want the path of least resistance

If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	C	Runs well	202.2 tok/s	522 ms	660K
Coding	C	Runs well	202.2 tok/s	958 ms	660K
Agentic Coding	C	Runs well	202.2 tok/s	1393 ms	660K
Reasoning	C	Runs well	202.2 tok/s	1132 ms	660K
RAG	C	Runs well	202.2 tok/s	1741 ms	660K

Quantization options

How Codestral 21B Pruned i1 (21B params) fits at each quantization level on Gaudi 3 128GB (128.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	8.2 GB	Low	D38
Q3_K_S	3	10.3 GB	Low	D38
NVFP4	4

Get started

Copy-paste commands to run Codestral 21B Pruned i1 on your machine.

Run

lms load hf-mradermacher--codestral-21b-pruned-i1-gguf && lms server start

Frequently asked questions

See all results for Gaudi 3 128GB See all hardware for Codestral 21B Pruned i1

URL: https://willitrunai.com/can-run/hf-mradermacher--codestral-21b-pruned-i1-gguf-on-gaudi-3-128gb

⇱